Surrogate optimization of variational quantum circuits
AI Breakdown
Get a structured breakdown of this paper — what it's about, the core idea, and key takeaways for the field.
Abstract
Significance Optimization on quantum hardware is required in many applications, including Hamiltonian simulation to quantum machine learning; this entails interesting problems that must be addressed both for noisy near-term and fault-tolerant hardware. We developed a method leveraging classical computing to accelerate optimization algorithms, which can vastly reduce the required quantum resources. We use the surrogate optimization technique; we learn approximate cost function landscapes utilizing cutting-edge classical simulation. We apply our approach to state preparation for chemical and condensed matter, showing speedup in all cases. We demonstrate we can use high-performance classical computing to better understand quantum advantage for optimization acceleration on quantum hardware.